摘要
针对日益增长的全球导航地图数据获取对众源地理信息的质量评估需求,选取全球覆盖率最高的开放街道地图OSM开展数据质量评估方法研究,提出一种可利用矢量路网数据和卫星遥感影像作为参考的实体级数据质量评估框架,将评估数据和参考数据进行自动匹配、融合和比对分析,从数据完整性、专题精度和定位精度3个方面对OSM路网数据进行质量评估。选择两个德国城市分别使用商业导航数据和高分辨率卫星遥感影像作为参考数据进行实验。实验结果表明:该方法不仅改进了传统的基于网格划分的质量评估方法,实现了在道路实体粒度上对OSM数据的有参质量评估,还引入了卫星遥感影像作为评估参考数据,方便了评估框架在缺乏矢量路网参考数据的地区应用,该方法可为众源数据质量审核与应用提供一种适用范围更广、自动化程度更高的实体级技术方案,有助于改变当前以人工作业编辑为主的导航地图数据质检模式。
In view of the growing need of quality assessment of crowdsourced geographic information for data acquisition in making global navigation map,OpenStreetMap(OSM);which has the highest coverage in the world,is selected to carry out the research of data quality assessment method,and an entity-level quality assessment framework is proposed for OSM road network data through utilizing vector road network data and satellite remote sensing imagery as reference data,in which automatic matching,conflating,comparing,and analyzing for the target crowdsourced data and the reference data with quality indicators of data completeness,thematic accuracy,and positioning accuracy.Two German cities are selected as experimental areas,and commercial navigation data and high-resolution satellite remote sensing images are used as reference data,respectively.The experimental results show that the proposed method not only improves the traditional grid-based quality assessment and realizes the reference-based quality assessment of OSM data at the road entity granularity,but also introduces satellite remote sensing images as the assessment reference data,which facilitates the application of the assessment framework in the regions lacking vector reference data of road network.And it can provide a more applicable and more automatic entity-level technical solution for quality auditing and editing of crowdsourced data,which helps to improve the current practices of navigation map data acquisition that heavily relies on labor-intensive quality inspection.
作者
杨剑
张猛
方立
贾奋励
周高静
张静茹
杨铭
侯洋
YANG Jian;ZHANG Meng;FANG Li;JIA Fenli;ZHOU Gaojing;ZHANG Jingru;YANG Ming;HOU Yang(Information Engineering University,Zhengzhou 450001,China;School of Human Settlements and Civil Engineering Xi'an Jiaotong University,Xi'an 712000,China;Quanzhou Institute of Equipment Manufacturing Haixi Institute Chinese Academy of Sciences,Quanzhou 362216,China)
出处
《测绘科学技术学报》
2024年第5期526-533,共8页
Journal of Geomatics Science and Technology
基金
国家自然科学基金重点项目(42130112)
国家自然科学基金青年项目(41901335)
国家重点研发计划项目(2017YFB-0503500)
智慧地球重点实验室基金资助项目(KF2023ZD04-02)。
关键词
众源地理信息
质量评估
数据匹配
数据融合
开放街道地图
crowdsourced geographic information
quality assessment
data matching
data conflation
Open-StreetMap
